Evolving notes, images and sounds by Luis Apiolaza

Page 4 of 67

Keeping track of my links

I have been using internet since 1993, which means thousands of browsed sites, broken links, storing and losing information for over three decades. One obvious point of the exercise is that every time I have relied on someone else’s system I have ended up losing lots of information., Twitter, etc. have consumed my data and time without an ability to maintain a good archive of my information. The only data that has remained is the one I have personally stored under my own system/payment. I have been a slow learner in this respect, so I have started another section of this site, Aleph, just to store bits and pieces of information I am collecting while browsing.

The first post in Aleph briefly documents the rationale; actually it just links to Cory Doctorow’s post to that effect.

Monopoly money vs real money

“Tree breeding has added 2 Billion dollars to the forest industry” said the presenter during a seminar.

Two billion? With a B? How come we struggle to get funding for projects then—I asked myself.

There are two testing cultures in forestry: the inventory/modeller crowd and the breeding crowd. The inventory crowd often relies on multiple-tree plots over a given area. Plots can be rectangular, circular, defined by prism, etc. The breeding crowd tends to use single-tree plots, because they (including myself) are testing many genotypes and it is more statistically efficient to use plots defined by a single individual.

Breeders use a selection index that gives a dollar value for each individual, while competing against a mix of genotypes (remember single-tree plots?). We would like to extend those results to inventory level and multiply the values by hundreds of thousands of hectares, there is a correlation with area-based performance, but not perfect.

Don’t get me wrong, I work in breeding. However, the selection values are Monopoly money until we get realised genetic gain validated by inventory plots in Real money. The two testing cultures have to match and forest valuation go up by $2 Billion before making a claim like that.

Teaching tsunami

One day you are all happy and relaxed, with plenty of time for cooking dinner of any kind. Three course dinner? No problem! Then, one day, just around the corner comes the start of semester. It is a bit like when one is distracted at the beach, waving hello to a friend or loved one and a sneaky wave tackles us and runs us over without any consideration.

On the plus side, I am usually looking for links that could be interesting to first year forestry students. Some times they are about forestry related news, but others are nice implementations of data visualizations that tell an interesting story. That’s when I came across this interesting clustering of New York City neighbourhoods grouped by similarity of proportions of street-tree species, done by Kieran Healy.

Of course one thing leads to the next until I got to this great New York City Tree Map in which one can get information for nearly 3/4 of a million individual trees. Enough to motivate some students. Zoom in until finding a tree you like.

Neighbourhood clustering by tree species

Old dog, etc

Yesterday I attended an introduction to “Artificial Intelligence in Forestry”* workshop at the School of Forestry, University of Canterbury. There were plenty of industry people, including some of I taught “a few” years ago.

I am mostly a quantitative genetics/stats person, but I keep my eyes open for anything that could be helpful dealing with my genetics trials, and have dabbled on using spatial things, LiDAR, etc. As I have pointed out in the past: 1. I am a sucker for entertaining problems and 2. old dogs can learn, albeit more slowly than 30 years ago, new tricks.

It was a nice opportunity to dip my toes in applications of machine learning for image analysis (detection and classification), mostly using ArcGis Pro and a short exercise using R. Funnily enough, my biggest barrier in the workshop was dealing with Windows + ArcGis. I am a Mac and Linux person, generally dealing with coding statistical analyses writing commands in text files. Most of my GIS related work is done via R (which I’ve been using since 1997) and any interactive work with maps using QGIS. Finding anything with point and click took me ages, until we got to the R part, which was hard for most people but kind of easy for yours truly.

Most importantly, having a chat with people and getting ideas to try with my data were the coolest part of the day. Thanks to Vega Xu, Justing Morgenroth and Ning Ye for organising the course!

*Confession: I dislike attaching the AI label to everything these days. Perhaps a better name would have been “Machine Learning in Forestry”. Small detail.

Classroom view from the back.
« Older posts Newer posts »

© 2024 Palimpsest

Theme by Anders NorenUp ↑