Poster nightmares

This week our university is having a symposium on all the research that occurs within our walls. I've been involved in producing three posters:

Poster 1: grassland research and the relationship between plant diversity and bird abundance and diversity (displayed Tuesday)

Poster 2: how acorns differ (or not) their chemistry over latitudes and how weevils that infect acorns also vary over latitudes (displayed Tuesday) 

Poster 3: how urbanization affects dispersal distance and the proportion of acorns that are eaten versus cached (displayed tonight) 

On Tuesday night, we put our two posters up and looked at them with pride. We used mixed model regression, multivariate regression, fancy method. Then the other posters from other departments went up - these were from business and nursing. I smiled smugly on the inside with our fancy posters. 

Then overnight it hit me. Our posters were terrible. TERRIBLE. The point of this exercise was to inform other departments of what we do. In that way, I can't imagine that a nurse was interested in our R-squares or our literature cited or the multivariate output. In neither poster did we explain why any of this is important for either society or science or both. In none of these posters do explain some of the possible really interesting implications.  

Also, our methods were all words. ZZZZZZZ. Is anyone going to read it? Probably not. 

So I'm scrambling to change the poster that's put up today. Let's see how it goes. 

On deck for the week 3/19/17

You'd think with four days off at work I'd be ahead. Never underestimate my ability to waste time. Worst part: did nothing with the puppy.

Things on my agenda for the week coming up

1. Meeting with provost about study abroad
2. Meet with two students about study abroat and 18 advisees
3. Finish two reviews (another one roled in yesterday... ugh) - one for a journal and other a book chapter that's essentially code
4. Give biostatistics exam
5. Hone biostatistics lab (manual is now 150 pages.... probably 1/4 the length it needs to be)
6. Hone biostatistics notes (just passed 100 pages.. probably 1/8 the length it needs to be)
7.  Think about how to fund a sabbatical in the tropics or Wales - yes, those are my choices
8. Figure out how to get 4 undergrads to the Mid-Atlantic Ecological Society of America meeting at Stockton State
9. Give a presentation to the alumni about study abroad
10.  Work on the Georgia paper 
11. Give independent research kids something to do

Habitat relationships in complex landscapes

I'm working on some old data from Columbus, GA that I should have published years ago. So it goes.  

This is something that has been bugging me for over a decade.  If you look at the relationship below which shows how native bird diversity changes with urbanization. 

Urbanization is quantified as the amount of impervious surface a km around where they were counted. The data are transformed but the whole scale goes from 0% to 100% (just take the sine of any of those numbers to transform back into a percent). 

Note that the relationship is positive from 0 to 20% and negative >20% urbanization. The post 20% is easy to understand: as urbanization comes to dominate a landscape, native birds find fewer habitats. 

Less than 20% there is little urbanization. That's easy to get. But what is there? In a real complex landscape, where there is not city there can be anything. It can be completely forested, agriculture, suburban, a mix of all the above. I think that explains all the noise on the low-urbanization side. 

I think what's lacking - in any statistical approach that I'm aware of - is how to model with factors whose effect can vary with their values. So in this case, I bet species richness increases with urbanization at first because a little bit of urbanization adds some birds associated with urban habitats but there's still enough natural habitat to have those birds persist but as urbanization increases the natural habitat gives way and birds are lost. 

So I need to invent a statistical method.