Posted on June 27, 2013 by James Douglas
Today was the first of two days dedicated to pulling our projects together. The more I thought about the toolset I had spent so much time working on, the more I came to realize it was relatively useless in providing helpful information for calculating water balance. The numerous scale issues, a complete lack of vegetative inputs, my inability to calibrate the solar radiation maps, and a lack of understanding of the outputs under further scrutiny left me feeling like I was back at square one! It was pretty frustrating, but I started to put together some preliminary maps so that I could at least talk about the data that I had collected, the inputs used, and if worst came to worst at least detail my failures.
During the morning I also retrieved (and by that I mean Emily- thanks!) a levelogger that I had placed in the stream (jk, that was Emily again) a few days back. The logger recorded data on stage, temperature, and conductivity. Since we had actually gotten some significant rain over this period, I was curious to see what the results held. It was nice to be able to actually make some progress in dealing with real field data, although I am not sure how helpful it will be in the grand scheme of things. One interesting thing I pulled out of the data was an inverse relationship between stage and conductivity during spikes in flow, which although isn’t a result that is overly surprising, is something I can at least talk about in my presentation. I also sat down and crunched the numbers on the discharge readings that I took yesterday, and it was nice to get some quantitative data organized and somewhat processed. I also made an attempt to correct the data for baseflow conditions by comparing the velocity reading from the thalweg from yesterday with an old reading I took last week while attempting to get elevation data. Although I don’t know accurate the adjustment is, I feel like it gets me as close as I’m gonna get.
I managed to get Dr. Wechsler free for a while and discussed my frustrations. After some exploration of the model and its confusing and overly simplistic outputs, her advice was to approach my project from the perspective of explaining the hydrology to the best of my ability, and to talk about the shortcomings of the model, the existing data, and provide recommendations for future study. Although this was a disappointing conclusion to reach, I realized that given the time constraints this was at least a viable and reasonable alternative, and I could leave the REU with the new knowledge I had gained about GIS and research in general.
I continued (trying) to focus on the project until dinner, and after dinner I learned that Dr. Becker would be paying us all a visit (yay!). He made his rounds and when he got to me, we discussed the limitations of the model, and he gave me some helpful tips in making use of the discharge data. We used a combination of ArcGIS tools, precip data, and mass balance calculations to estimate the difference between the observed discharge and what a theoretical discharge would be ignoring infiltration, interception, and other real world factors. This helped us start to quantify how much of the precipitation was making it to the stream as baseflow. Turns out only about 12-14% of the average precipitation is making it to the stream as measurable flow. I also managed to find a paper where mass balance calculations had been run on the watersheds on the windward side of Oahu (score!). This allowed me to fall back on some literature for what might be expected for the relative breakdown of water movement through the watershed- runoff, gw recharge, and evapotranspiration. This was quite a breakthrough, and so I ran through some calculations that I could actually present. It’s been quite a long day, and honestly I’m about ready to present this thing, but there’s still a lot of work to do tomorrow… but now for sleeeeeeeeeeeep