Wrapping up the Summer and a Look Ahead

My research for the summer is now coming to a close, and I have now created an ocean of data to swim through. And, I’m finding, the process of research isn’t so much about answering questions as much as asking questions. Which can be really frustrating at times, when the questions really pile up. Learning to ask one simple, answerable question at a time is something I am very much still working on.

I have also learned about working on a long-term project. As the Jamestown water quality project has been the focus of 5 previous undergraduate theses, I have watched through the literature as new questions get asked and answered. I also have watched as errors (sometimes from a data source’s error and not researcher’s mistake) have led to mistaken conclusions. At least some of what I am doing is to correct these, and to provide data to support some conclusions that were not fully supported. In short, research is a process that can be messy and frustrating, but has a special kind of reward all its own. For me, satisfactorily answering one of those simple questions is akin to the relief found by scratching an itch.

Looking Ahead

The past few days I have been playing around with some USGS groundwater modeling software, Modflow 2005, and a graphical interface called ModelMuse. I plan to explore using the MT3DMS add-on for including water salinity transport. These will hopefully allow me to make models that will mirror the situation we see today and then to predict how the environment of the early colony would have affected things.

Calibrating models is notoriously tricky. There are an infinite number of variables involved, but the ones that are probably the most important here are: James River elevation, Pitch and Tar Swamp elevation, aquifer properties (hydraulic conductivity and specific storage), precipitation input, and evapotranspiration. I’m looking into ways to compare the head and salinity we find in our observation wells to what the model predicts at a given time. I expect this to take months to do.

I have gotten a start on it, mainly just figuring out how to use the program. Here are some of my simulations so far. I recommend clicking on the images (twice) to view them full-size because WordPress won’t let me make them any bigger.

Jtown Map

Just as a refresher, here is an aerial photo of the site. Observation wells are labeled

Terrain-For Blog

This is the same site in ModelMuse, with the elevations approximated (5x vertical exaggeration). The fort and monument are on high ground (~4m elevation NAVD88).

Boundary Conditions

I set the swamp and river to their exact elevations on this day. I put the swamp as extending 1m below the surface and the river 5m below the surface. For this model, they are considered constant head boundaries, so they don’t change. In reality, they change a great amount. Making them change through time is something I will do later.

Top View- For Blog

This is the top view of the running the simulation for 5 years. I did it that long to make sure that it is completely in equilibrium. I input the hydraulic conductivity values that I measured in my slug testing. The model solves for head at each cell in 3 dimensions (it is 10m deep). The legend refers to head, so water flows from the swamp to the river.

Head Side On- For Blog

Here is a side-on view that shows how it is 3D. Notice that some of the cells are empty. Because this is an unconfined aquifer, the water table doesn’t intersect with the land surface except at water bodies (swamp and river).

Test2

A simulation run that generates contour lines, rather than a color map. It also gives you an idea of what ModelMuse looks like.

I have a lot of learning to do about how to use all of this, and what it can tell us about the water quality of the Jamestown Colony. But at least it looks pretty.

Thanks for reading. I’m signing off for the summer. Here’s to more progress in the fall!