Ok, so it’s the end of my 8th week of summer research and things seem like they’re finally starting to come together a little bit. Last week, Max Cunningham and I spent our last days in the field exploring variations in rock strength over the longitudinal profile of Renick Run. We spent 14 total days in the field and ended up with thousands of rock strength measurements for 19 cross sections. I am happy because I feel like we’ve accomplished something pretty rare; not many geomorphologists study any one stream as closely as I feel like we’ve studied Renick Run. Also, the initial data look encouraging. Though the results are noisy (and more to the point are Max’s research focus, not mine), there are some initial patterns that appear to show systematic changes in rock strength with cross-sectional channel geometry and distance along the longitudinal profile. But I’ll leave it to Max to discuss those in depth.
The one major downside about spending all of this time out in the field collecting rock strength and channel geometry measurements is that I have had much less time than planned to focus on my numerical modeling. It doesn’t look like I will be able to accomplish my initial goal of modeling channel profile evolution this summer. However, the news isn’t all bad! As I discussed in my last blog post, I’ve spent my time in the lab exploring channel responses to one-time changes in the baselevel lowering rate. In my last post, I said that I was working towards comparing the times that channels took to re-equilibrate after these changes. Now I have significant amounts of data to that end that are yielding interesting (if confusing) conclusions. I ran two major sets of models in MATLAB, one of which allowed the channel to freely weather and one that turned weathering off to explore channel response in the absence of rock weathering. Each set consisted of 25 model runs, each of which tested a different factor of change in the baselevel lowering rate. Here, I present a small amount of the data I’ve extracted using channel slope as a test for equilibrium conditions in the channel.
This first graph shows the time that each modeled channel took to re-equilibrate (Y-axis) after undergoing a given change in the baselevel lowering rate (X-axis). The results look similar to what we would predict: the higher the magnitude (in either direction) of the change in the baselevel lowering rate, the longer the channel takes to reach its new equilibrium. When looking at this plot it is important to remember that weathering was turned OFF in the model, meaning that the rock strength is assumed to be constant throughout the cross-section.
Now for the interesting part. This next plot is exactly the same except that it shows model runs that allowed the rock in the channel to freely weather. The results are very different from anything I predicted, and definitely will warrant further exploration in my last week of summer research.
For x=1 to x=50, this plot shows times to equilibrium that are extremely close (though slightly lower) to the times returned by the no weathering models. However, for x=.01 to x=1, we see completely different values and trends that are not readily explained by our knowledge of the weathering model.
My next steps are to a) Plot the width-depth ratios and vertical erosion rates of these model runs to see if they follow the same patterns and to b) explore the raw data for rock strength, width, and depth from the model to see if this odd pattern on the left side of the second graph can be explained. Time will tell!