Year: 2020 Page 1 of 3

We just received notice that our first article titled Camera-based Water Stage and Discharge Prediction with Machine Learning has been accepted for peer review by the Hydrology and Earth System Science .Journal (HESS). It is an online journal that selects two or more official reviewers, but is also left open for public review and comment for two months. This does not mean the article has been accepted for publication, but it is the first step in the process with the hope that we can get it accepted.

In the meantime, we have defined the research plan for the next article which will partly a replication study and partly the development of tools that make it easier for others to duplicate our work. According to my defined “program of study,” I have to write three of these articles to get to the point where I can take start writing my dissertation. In addition to that, I have to take 5-7 additional courses to combine with my Masters degree courses before I have met the minimum requirements of the plan. It seems like there is a ton left to do, but it also seems like we are off to a good start. We will find out if that is true or not by seeing if this paper gets accepted.

This is an amazing story. I have been pretty self-congratulatory about starting a STEM PhD at age 64 with the hope that I can get it done by the time I am 70. A guy named Lucio Chiquito from Medellin, Colombia, just submitted his dissertation to University of Manchester in England at age 104. I thought, well, he probably got it in Sociology or Spanish or History, but no, he got it for work on a tough mathematical characterization of the flow of water in rivers. Guess what I am studying? The study of the flow of water in rivers. His dissertation is a harder version of what I am doing. Notice his little helper down in the bottom right of the photograph. The thing that I liked the most about the guy was, when they asked him what he was going to do now, he told them he wanted to work on perfection his English and German–continuing to learn. Beautiful stuff.

We heard back from the editor for our first journal article. Rather than send it on to the reviewers, he got back with some recommendations he wanted us to make before he did that. We worked feverishly to get that done, asked for an additional week, got permission for two additional weeks, and the paper was canceled. We think this was due to the fact that we got the permission with plenty of time (we are in Pacific time), but it was after the European editorial office had already closed, so we do not think it will be a problem.

We could have gotten by just removing some verbiage the editor did not like (we are agnostic about that part and do not care so we removed it), modifying the format of the equations (this, we believe, is a matter of fashion–one of our authors is a theoretical mathematician and thought it was silly, but again, we do not care so we changed it), he wanted us to change the x-axis labels of some of the graphs (he was definitely right on this one), and he wanted us to compare our results to a specific kind of upper and lower benchmarks. The last item was the cause for the most major revisions and think the paper is a lot better for having done it. We had to add a graph and a table, some citations, and some further analysis, but we think it helped the paper a lot.

I started my first class. There are thirty students from UNL (5) and different schools in Brazil (25). So far, I am enjoying it a lot and it makes sense. That may change, but so far I am not lost (my worst fear) nor am I fearful that I am not going to learn anything. I think I am going to learn a ton and that it will directly apply to my doctoral research. I will get to learn a new tool.

This semester, I am taking a course for my retirement PhD project titled “Groundwater Modeling through Time Series Analysis.” It is only a one-credit “Special Topics” class, but it looks very interesting and a good way to get my feet wet back in the (remote) classroom. The course is to be taught by my PhD adviser, Troy and one of his compatriots in Brazil. There will be a sum total of three lecture sessions and three working sessions to do projects with a software package called HydroSight. The material looks great, but the program depends on the MatLab runtime so that did not please me. Anything that depends on the MatLab runtime is a disaster and a tragedy, but that is a rant for another day. I am really looking forward to the course. It starts on Tuesday and I have permission from my day job to sit in on the class in the middle of the day.

Last week, I was able to submit the first refereed journal article for my retirement PhD research. Right now we are waiting to see whether it will be accepted and to for the formation of a review committee at the Hydrology and Earth System Sciences (HESS) journal. We think it is a pretty good article about how to create machine learning models with image features from annotated with USGS stage (water level) and discharge (stream flow) sensor measurements to fill gaps in the data when there is equipment or funding failure. We actually got some pretty amazing results. The images were not taken in a way that is normal for scientific or engineering purposes, but from a documentary project (Platte Basin Timelapse project). Here is a graph from an earlier post.

This shows the observed vs. predicted graphs for the data gap filling. The middle (2016) year did not predict so well in the preliminary effort, but we addressed that and made better predictions in the paper.

A poster based on the research we performed for our first article has been accepted for presentation at the American Geological Union (AGU) that was supposed to be in San Francisco in December, but that now will be online. That if a pretty cool deal and we think we have a pretty cool poster (in progress). My adviser, Troy, sent me a note before he went away camping for a long weekend to the effect that poster preparation is not done in the normal way–creation of the poster in LaTeX or PowerPoint, then printing it out in a large format. Because it is an online venue they want everyone to create their posters with online software they provide. I am assuming this is do to their incompetence because it will be a big hassle for everyone and almost certainly decrease the quality of the presentations.

The second draft of the first journal article for my PhD research is complete and out for review. I have moved on to start writing the software for the second article. The first part of that is improvement and refactoring of the old GaugeCam project. I finished the refactoring part and am moving on to the improvements. Click on the image to see a video of what the program does so far. I am having a lot of fun with it. We hope to have the program ready for full use by February or March of 2021. At that point I can start writing the second paper.

Well, I have submitted the graphs for my first paper to my adviser and the other authors. Hopefully, they are all now complete, but I am pretty green at this and it seems like changes are always found. Whatever the case, I am now half-way through the first draft of the paper. Tonight I hope to get the figures placed into the document and maybe one more paragraph written. Who knows, I might be able to finish my first draft by the weekend.

Last night I joined the American Geophysical Union so I could submit an abstract for consideration as a presentation at their annual conference. I have no idea whether or not it will be accepted, but it seems like I have finally joined the fray. The conference is not until December and we will not know whether or not our abstract has been accepted until early October. In the meantime, I have a boat load of unorganized data, some graphs and a LaTeX document started with a large and growing bibliography. I cannot believe I am saying this, but I am honestly enjoying the process. We have a great team of authors from three different institutions and momentum for the follow-on paper if this one ever gets accepted. I will keep you posted.