Find recent updates from me here!

The Last Chapter

posted Oct 3, 2016, 9:02 AM by Austin Milt   [ updated Oct 3, 2016, 9:03 AM ]

An eery title for an eery topic: the last of my dissertation chapters is finally published in full, and (probably for the 5th time now) my PhD feels complete. Head over to my CV page to get links to and pdfs of all my chapters. I must say, it was a long and usually unpleasant experience getting my chapters published, but the works are definitely improved and I'm proud to have them out! A celebratory sleepy red fox is in order!

Featured in a Textbook

posted Sep 26, 2016, 8:05 AM by Austin Milt   [ updated Sep 8, 2017, 6:15 AM ]

I just learned that some of my PhD work has gotten a little spotlight in Craig Groves' and Eddie Game's new book: Conservation Planning: Informed Decisions for a Healthier Planet. I actually haven't seen the final one-pager yet (my copy is shipping), but knowing Craig and Eddie and the quality of their work, it's a pretty huge honor.  It's good to see others found my work interesting from a conservation planning perspective. Of course the real measure of success is whether practitioners and/or industry use it!

Weekly update 67

posted Nov 6, 2015, 8:12 AM by Austin Milt   [ updated Nov 6, 2015, 8:12 AM ]

We had an interesting lab meeting this week. Alex Orellana came in from the UW Design Lab to help us think about the design of our academic figures, illustrations, and diagrams. We covered a lot of stuff I have heard and practice, e.g. number and types of fonts, whitespace, contrast, symmetry, etc. 

There were also some things that I hadnt heard about before. For example, Alex suggested that we try projecting and sketching by hand to create interesting line-art (using illustrations of animals as an example), then scan those into a computer and convert to vector graphics. It's an interesting concept. I would have tried to use software to do the whole process, but he pointed out the software renderings usually look... artificial.

One point that came up a few times in our discussion was the standard lack of emphasis on design in scientific illustrations. Because journals have tight restrictions on the use of space, we often try to pack as much information in a tiny figure as we can. And to some extent that's necessary. But in the examples that Alex showed us, when juxtaposed against our own figures, is that modern design simplifies to the extreme, conveying only a couple types of information in a larger figure rather than packing it all in. I think the result is that the most relevant messages we want to convey are done so more strongly and immediately. It's an important consideration. Journals are more and more moving toward short, concise, general audience formats. The explosion of scientific articles we're expected to peruse also means we have less time to spend on each one. So if we want our articles to keep up with the times, we have to design the graphics to be punchy and convey core messages quickly.

Weekly update 66

posted Oct 30, 2015, 8:54 AM by Austin Milt   [ updated Oct 30, 2015, 8:54 AM ]

I've (re)discovered this week that reproducing and interpreting others' code has some pretty strong pros and cons. Pro: someone else has already done the work of designing and implementing an algorithm. Con: designing and implementing algorithms (that work) is mostly what I enjoy about coding; it's problem solving on a small scale. Pro: someone else understands what you're doing and can help when you get stuck. Con: sometimes you have to depend on others when you're stuck and they have busy schedules and other priorities, which means others can significantly slow down your progress. Pro: variables already have names. Con: sometimes those names are hopelessly confusing. Pro: you get to learn while you read. Con: but only if there's good enough documentation to teach you.

Weekly update 65

posted Oct 2, 2015, 8:12 AM by Austin Milt   [ updated Oct 2, 2015, 8:12 AM ]

This week I attended the Healing Our Waters conference in Chicago. Much like TNC's All Science Meeting, this conference had a lot of practitioner oriented presentations, which is an interesting change from the typical academic conference. Presentations tend to be case-study focused and rarely have a head in the clouds message. Unlike the All Science Meeting, HOW brings together tens of regional practitioner organizations ranging from watershed level land trusts up to regional or larger multi-purpose organizations. As a result, the attendees ranged from jeans and wrinkled button-up (me) types up to the three-piece suit organization heads, and the way each person thinks and acts along that spectrum is marked and interesting.

My favorite session was about systematic conservation planning in Saginaw Bay (Wednesday late morning session). It stood out for two reasons. First, rarely if ever have I seen such comprehensive planning in an applied setting. The group integrated preference surveys, biological data, hydrology modeling, and site selection to come up with priority watersheds feeding into Saginaw Bay. They had nearly every piece of the puzzle at least partially addressed, and I was impressed. Second, the group was totally focused on Saginaw Bay, when their methods and potential for a broader effect were obvious. They traded off practicality (do something well at a smaller scale) for breadth. We academics in conservation planning are often encouraged to look further and state how our little case study will affect the whole world and, sometimes, that prevents us from getting anything done.

Weekly update 64

posted Sep 25, 2015, 8:16 AM by Austin Milt   [ updated Sep 25, 2015, 8:16 AM ]

Two especially good things have happened this week. First, I finally have the go-ahead to submit my third chapter of my dissertation for publication, months after I had planned on it (of course). This is supposed to be the big splash chapter, so my fingers are crossed it gets favorable reviews. Second, I've now got a couple project ideas down on (virtual) paper and am waiting from my boss to hear what he thinks about them. The projects come down to two questions (background/context omitted):

What is the benefit for small scale connectivity restoration to consider downstream options? I.e. if I'm interested in the quality/accessibility of my local stream, when/how/how much should I consider instead of investing downstream of my local stream rather than directly in it?

The other project is: What is the cost effectiveness gain in centralizing stream connectivity restoration and lamprey control decisions? This is supposed to be a large scale analysis that tries to address the current separation of interacting goals for conservation in the Great Lakes.

These are by no means set in stone, but it's exciting to at least have something to start working towards, even if I end up with a totally new destination.

Today I am going to try to get my feet wet working on some GAMS code. Let's see if I can read and write this stuff.

Weekly update 63

posted Sep 18, 2015, 8:23 AM by Austin Milt   [ updated Sep 18, 2015, 8:23 AM ]

Not surprisingly, there's more underlying my pending work here at the CFL than at first appeared. Data are data, and models are models, but sometimes there's more or less and further along or behind than you thought. I'm still working out just what it is I need to prioritize in my tasks here, but it seems inevitable I will come out with some neat research ideas. 

One thing I did this week was play around with the DST (Fishwerks) being developed by my team, including some folks over at the Wisconsin Institute for Discovery. It's a pretty cool interface and tool with some quirks to be worked out. Probably my favorite thing about it is that you can run some pretty advanced applied optimizations on a dedicated cluster without needing special certification (note this is not an endorsement to overload the system).

Weekly update 62

posted Sep 11, 2015, 9:14 AM by Austin Milt   [ updated Sep 11, 2015, 9:15 AM ]

Week 2 down, and things are starting to heat up. I've been reading umpteen papers and have now met with several of the core folks on the project, including Matt Diebel, Allison Moody, Michael Ferris, Tom Neeson, and Jesse O'Hanley. It looks like there's a few things for me to be working on, all of which involve updating and integrating cost regressions into optimization models. On one front, there's new data for barrier removal/upgrade cost models. On another front, there's refining and integrating a lampricide application cost into the optimization models. This will take some learning on my part, notably: what are all the input data and why these?; reading/writing GAMS code; doing some GIS; and other less interesting tasks.

Next week I hope to start pinning down some research questions so all this work I'm about to embark on has a concrete end-goal!

Weekly update 61

posted Sep 4, 2015, 8:01 AM by Austin Milt   [ updated Sep 4, 2015, 8:01 AM ]

It's Friday of my first official week as a postdoc. I've got a desk, a computer, a UW ID and email, and a laundry list of things to read and people to meet. Things will start in earnest next week, as I plan to spend a lot of time reading and meeting with some core folks on the team. I'm excited to get going on this next part of my academic career!

Weekly update 60

posted Jun 19, 2015, 9:12 AM by Austin Milt   [ updated Jun 19, 2015, 9:12 AM ]

I've been juggling paperwork and manuscript revisions over the past several weeks. I am happy to say that the paperwork is for a postdoc I am starting September 1 in the McIntyre lab at the Center for Limnology in Madison, WI. It's a two-year position working on freshwater connectivity in the Great Lakes Basin. While it's a big change topically, it sticks close to home in regards to the type of conservation problems I like working on and its immediate relevancy for conservation.

1-10 of 71